Feedback-Control Operators for Evolutionary Multiobjective Optimization

Authors:
Ricardo H. Takahashi;Frederico G. Guimarães;Elizabeth F. Wanner;Eduardo G. Carrano
Affiliations:
Department of Mathematics, Universidade Federal de Minas Gerais, Brazil;Department of Computer Science, Universidade Federal de Ouro Preto, Brazil;Department of Mathematics, Universidade Federal de Ouro Preto, Brazil;Centro Federal de Educação Tecnológica de Minas Gerais, Brazil
Venue:
EMO '09 Proceedings of the 5th International Conference on Evolutionary Multi-Criterion Optimization
Year:
2009

Citing 8
Cited 1

Modern Control Engineering

Modern Control Engineering
Genetic Algorithms for Multiobjective Optimization: FormulationDiscussion and Generalization

Proceedings of the 5th International Conference on Genetic Algorithms
A Fast Elitist Non-dominated Sorting Genetic Algorithm for Multi-objective Optimisation: NSGA-II

PPSN VI Proceedings of the 6th International Conference on Parallel Problem Solving from Nature
Estimation of pareto sets in the mixed H2/H∞ control problem

International Journal of Systems Science
An overview of evolutionary algorithms in multiobjective optimization

Evolutionary Computation
Local search with quadratic approximations into memetic algorithms for optimization with multiple criteria

Evolutionary Computation
Local models--an approach to distributed multi-objective optimization

Computational Optimization and Applications
Performance assessment of multiobjective optimizers: an analysis and review

IEEE Transactions on Evolutionary Computation

On a stochastic differential equation approach for multiobjective optimization up to pareto-criticality

EMO'11 Proceedings of the 6th international conference on Evolutionary multi-criterion optimization

Quantified Score

Hi-index	0.00

Visualization

Abstract

New operators for Multi-Objective Evolutionary Algorithms (MOEA's) are presented here, including one archive-set reduction procedure and two mutation operators, one of them to be applied on the population and the other one on the archive set. Such operators are based on the assignment of "spheres" to the points in the objective space, with the interpretation of a "representative region". The main contribution of this work is the employment of feedback control principles (PI control) within the archive-set reduction procedure and the archive-set mutation operator, in order to achieve a well-distributed Pareto-set solution sample. An example EMOA is presented, in order to illustrate the effect of the proposed operators. The dynamic effect of the feedback control scheme is shown to explain a high performance of this algorithm in the task of Pareto-set covering.